Effect of hip fracture on mortality in elderly women: The EPIDOS prospective study

Institut National de la Santé et de la Recherche Médicale (INSERM), Villejuif, France.
Journal of the American Geriatrics Society (Impact Factor: 4.22). 05/2004; 52(5):685-90. DOI: 10.1111/j.1532-5415.2004.52203.x
Source: PubMed

ABSTRACT To assess whether the occurrence of a hip fracture is associated with an increased risk of mortality even after taking into account age and prefracture health status and whether this increased risk of mortality persists beyond the first 6 months after the fracture.
A prospective study of risk factors for hip fracture.
Five French areas: Amiens, Lyon, Montpellier, Paris, and Toulouse.
The cohort consisted of 7,512 volunteer ambulatory women aged 75 and older who were recruited from voter registration lists. Women who had a history of hip fracture or bilateral hip replacement were excluded.
The baseline examination included a functional and clinical examination and a questionnaire on life style and treatments. Thereafter, women were followed every 4 months for 4 years to record the occurrence of fractures and deaths. A multivariable proportional hazards model was used to determine the association between hip fracture (treated as a time-dependent variable) and mortality, after adjustment for age and baseline health status.
During a mean+/-standard deviation follow-up of 3.9+/-0.9 years, 338 women had a first hip fracture, and their postfracture mortality rate was 112.4 per 1,000 woman-years, compared with 27.3 per 1,000 woman-years for the 6,115 women who did not have any fracture (P<.001). After adjusting for age and baseline health status, women with hip fracture were more than twice as likely to die (95% confidence interval (CI)=1.6-2.8). This increased risk appeared more pronounced in the first 6 months (relative risk (RR)=3.0, 95% CI=1.9-4.7) than after (RR=1.9, 95% CI=1.6-2.2) (P=.09).
In ambulatory elderly women, the occurrence of a hip fracture is associated with an increased risk of death, even after prefracture health status is taken into account. Although the effect of the fracture is stronger in the first 6 months, it persists for several years thereafter, which suggests that prevention of hip fracture and improved care after the fracture may contribute to increase life expectancy in addition to preserving quality of life.

  • [Show abstract] [Hide abstract]
    ABSTRACT: To identify to which degree patient-related factors (age, gender, place of residence, general health condition, comorbidity) and hospital-related factors (waiting time for surgery, type of surgery, in-hospital complications, length of stay) may predict 1-year mortality in elderly hip fracture patients in an orthogeriatric unit, to optimize treatment and care. In-hospital patient information was routinely collected by multidisciplinary staff and entered into a database. Information about mortality was obtained for 567 patients aged 65 and above. Multivariate logistic regression was performed. Overall mortality was 23.5%, but there was a large variation in mortality according to age, gender, comorbidity, and place of residence. Independent predictors of mortality were admittance from nursing home (risk ratio [RR] = 3.24, 95% confidence interval [CI] = 2.37-4.43 compared with home dwellers) and a higher American Society of Anesthesiologists (ASA) score (RR = 1.75 and 95% CI = 1.24-2.46, for ASA ≥ 3 compared with ASA ≤ 2). Male gender, increasing age, increasing number of comorbid conditions, and having fallen indoors were indicators, but not independent predictors, of higher mortality. Almost one fourth of older hip fracture patients in this unit died within a year. The most important predictor was admittance from nursing home, which was associated with comorbidity and frailty. More attention to patients from nursing homes is needed in the health care system.
    Journal of Aging and Health 09/2010; 22(8):1114-31. DOI:10.1177/0898264310378040 · 1.56 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Bone loss, a fracture risk factor, may play a role in post-fracture mortality. We found accelerated bone loss (≥1.31 % bone loss/year for women and ≥1.35 % bone loss/year for men) associated with 44-77 % increased mortality. It remains unclear whether bone loss is a marker or plays a role in mortality. Osteoporotic fractures are associated with increased mortality although the cause is unknown. Bone loss, a risk factor for osteoporotic fracture is also associated with increased mortality, but its role in mortality risk post-fracture is unclear. This study aimed to examine post-fracture mortality risk according to levels of bone loss. Community-dwelling participants aged 60+ from Dubbo Osteoporosis Epidemiology Study with incident fractures were followed from 1989 to 2011. Kaplan-Meier survival curves were constructed according to bone loss quartiles. Cox proportional hazard models were used to determine the effect of bone loss on mortality. There were 341 women and 106 men with ≥2 BMD measurements. The rate of bone loss was similar for women and men (women mean -0.79 %/year, highest bone loss quartile -1.31 %/year; men mean -0.74 %/year, highest quartile -1.35 %/year). Survival was lowest for the highest quartile of bone loss for women (p < 0.005) and men (p = 0.05). When analysed by fracture type, the association of bone loss with mortality was observed for vertebral (highest vs lower 3 quartiles of bone loss, women p = 0.03 and men p = 0.02) and non-hip non-vertebral fractures in women (p < 0.0001). Bone loss did not play an additional role in mortality risk following hip fractures. Importantly, overall, rapid bone loss was associated with 44-77 % increased mortality risk after multiple variable adjustment. Rapid bone loss was an independent predictor of post-fracture mortality risk in both women and men. The association of bone loss and post-fracture mortality was predominantly observed following vertebral fracture in both women and men and non-hip non-vertebral fracture in women. It remains to be determined whether bone loss is a marker or plays a role in the mortality associated with fractures.
  • [Show abstract] [Hide abstract]
    ABSTRACT: Mandibular cortical width (MCW) detected on panoramic radiographs may be useful for identifying postmenopausal women with osteoporosis. There is little known regarding whether alveolar bone loss (ABL) of the mandible detected on panoramic radiographs is a potentially accurate screening tool for osteoporosis in comparison with MCW. The purpose of this study was to evaluate whether ABL of the mandible on panoramic radiographs is useful for identifying femoral osteoporosis in postmenopausal women in comparison with MCW. Three hundred and fifty-four Japanese postmenopausal women (mean age+/-SD, 56.8+/-7.7 years) were recruited for this study. Femoral BMD was measured by dual energy X-ray absorptiometry. Panoramic radiographs were obtained to estimate ABL of the mandible and MCW. A multiple regression analysis revealed that femoral BMD was significantly associated with MCW (P<0.001), weight (P<0.001), age (P<0.001) and ABL of the mandible (P=0.029; adjusted r(2)=0.380). The area under the ROC curve (AUC) for identifying femoral osteoporosis was 0.609 [95% confidence interval (CI), 0.523-0.696] for ABL of the mandible and 0.779 (95% CI, 0.713-0.844) for MCW, respectively. AUC for ABL of the mandible indicated less accuracy. Our results suggest that ABL of the mandible on panoramic radiographs may not be useful for identifying postmenopausal women with femoral osteoporosis in comparison with MCW.
    Dentomaxillofacial Radiology 01/2007; 36(1):28-33. DOI:10.1259/dmfr/28366679 · 1.27 Impact Factor